US 12,468,894 B2
Using language model to generate recipe with refined content
Prithvishankar Srinivasan, Seattle, WA (US); Saurav Manchanda, Seattle, WA (US); Shih-Ting Lin, Santa Clara, CA (US); Shishir Kumar Prasad, Fremont, CA (US); Riddhima Sejpal, Dublin, CA (US); Luis Manrique, New York, NY (US); and Min Xie, Santa Clara, CA (US)
Assigned to Maplebear Inc., San Francisco, CA (US)
Filed by Maplebear Inc., San Francisco, CA (US)
Filed on Sep. 8, 2023, as Appl. No. 18/244,098.
Prior Publication US 2025/0086395 A1, Mar. 13, 2025
Int. Cl. G06F 40/30 (2020.01)
CPC G06F 40/30 (2020.01) 20 Claims
OG exemplary drawing
 
1. A method, performed at a computer system comprising a processor and a computer-readable medium, comprising:
generating a first prompt for a language model, the first prompt including a plurality of task requests for generating initial content for a recipe;
requesting the language model to generate, based on the first prompt, the initial content for the recipe;
generating a second prompt for the language model, the second prompt including the initial content for the recipe and contextual information about the recipe;
requesting the language model to generate, based on the second prompt, refined content for the recipe; and
storing, in a database of an online system, a version of the recipe having the refined content.